Clinical characterization of neck pain in migraine
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.227302
Source of Support: None, Conflict of Interest: None
Keywords: Cervical migraine, migraine, neck pain, postdrome, premonitory phase, trigger
Headache and neck pain are common complaints among patients visiting the Neurology outpatient department. It has been speculated, probably for more than a century, that neck pain is a common and prominent symptom of primary headache disorders, including migraine. Migraine is a common cause of primary headache disorder and has a 1-year prevalence rate of 12% in the general population worldwide (affecting 18% women and 6% men). In India, the age-standardized 1-year prevalence rate is 25.2% (10.6% definite, 14.6% probable migraine).
Neck pain and pericranial muscle tension or tenderness is often thought to be a unique feature of tension-type headache, given its hypothesized muscular etiology. On the other hand, as migraine has a putative neurovascular etiology, it is difficult to explain neck pain in migraine based on muscle tenderness. In a recent study, a 1-year prevalence rate of neck pain was higher in patients with primary headaches (85.7%) as compared to those without headache (68.4%). Among patients with primary headaches, 88% of patients suffering from tension-type headache and 76% of those suffering from migraine headache were associated with neck pain anytime during the attack.
Nowadays, most authors believe that neck pain along with pericranial muscle contraction or tenderness are quite common in migraineurs, and the prevalence is somewhat similar to what is observed in tension-type headache sufferers.
Sometimes neck pain is the triggering factor, and the myofascial pain in the neck is the initiation site of migraine attack. This neck pain, after an initial trigger, may last during all phases of a migraine attack. The exact pathophysiology behind this is not known. Some authors term this type of migraine attack as “migraine cervicale.” We prefer to call it “cervical migraine.”
In clinical practice, we often notice some patients stating that neck pain triggered their migraine. At times, they would state that the neck pain got exacerbated typically if the neck had been held in one position for lengthy periods during driving or working on a computer, and at times, it triggered the migraine attack. In some cases, the neck pain existed due to cervical spondylosis or neck trauma, and sometimes, this may precipitate the migraine attacks. In the present study, we evaluated the incidence, demographic profile, and clinical features of cervical migraine.
Aim of the study
This study was conducted with three aims in mind: (a) To study the incidence and demographic profile of neck pain associated with migraine; (b) to study the association of neck pain in different phases of migraine, particularly in an attempt to understand whether neck pain is essentially a part of the symptom complex of migraine or it also acted as a trigger to precipitate an attack of migraine; and, (c) to study the various clinical characteristics and features, as well as associated symptoms, among patients suffering from cervical migraine.
This study was conducted as a prospective observational study at Sanjay Gandhi Post Graduate Institute of Medical Sciences, Lucknow, India from March 2015 to October 2016. The patients were screened on their first visit to the outpatient clinic. They were subsequently followed up directly during their hospital visit and also telephonically, whenever required, for clinical detailing. Whenever necessary, the patients were advised to report about their recent attack semiology immediately by phone or other electronic media (such as fax, email, or video call). The history taking, clinical examination, and necessary investigations were carried out after obtaining consent from each patient (based upon the patient information datasheet used in our hospital).
The detailed demographic parameters were recorded for each patient such as name, age, gender, religion, address, education, occupation, socioeconomic status, daily activities, dietary patterns, and behavioral and social relationships.
Headache interviews were designed to define the characteristics of migraine according to the International Classification of Headache Disorders–3-beta (ICHD-3-β). Migraine events in patients were classified as: migraine without an aura or migraine with an aura, episodic migraine or chronic migraine, complications of migraine, or episodic syndromes associated with migraine. Associated migraine symptoms such as sensitivity to light, sound and smell, nausea, vomiting, vertigo, aggravation of pain by physical activity, cranial autonomic symptoms, medication-overuse headache, and any associated primary headaches (such as the presence of a tension-type headache) were also recorded.
Neck pain interviews
Neck pain is defined as a subjective unpleasant sensory experience or myofascial pain in the neck, particularly in the nuchal or upper cervical region (nuchalgia). Individuals were asked about the side (unilateral or bilateral), radiation (to the upper extremities, head, or back), quality of neck pain (stabbing, cramping, lancinating, pulsating, and numbing), as well as aggravating and relieving factors related to the neck pain. If the patient was diagnosed as suffering from migraine headache and had neck pain, the following questions were asked for each attack (a minimum of five attacks of migraine with neck pain were considered for inclusion in the study):
Inclusion and exclusion criteria
Patients of both genders, aged 14–65 years, diagnosed as having migraine according to the definition of the ICHD-3-β, and having neck pain anytime during the attack phase, were included in the study. They were required to be having a minimum of five migraine attacks associated with neck pain at the time of their first visit to the hospital.
The patients who had a history of background neck pain, with an aggravation of neck pain during the migraine phase, were considered as 'neck pain-triggered migraineurs'. On the other hand, the neck pain was thought to be 'a part of the clinical presentation of migraine' in those patients who had neck pain only during the migraine phase, without having any history of neck pain in their non-headache phase.
We considered both these groups of patients as those suffering from cervical migraine; and, patients suffering from cervicogenic migraine belonged to that subset of patients having cervical migraine, in whom neck pain triggered the migraine headache.
Patients with known brain lesions or suspected to be having one, those with a significant cervical pathology except cervical spondylosis, as well as those with a history of significant brain or cervical trauma or surgery, fibromyalgia, and any somatoform syndrome were excluded from the study.
Data were summarized using descriptive statistics, that included either the number or percentage of patients in each category, as well as the mean, standard deviation (SD), median, and range. Statistical analyses were performed using the Statistical Package for the Social Sciences (IBM) for Windows, version 20.0 (Java). Two-tailed P values were calculated and those less than 0.05 were considered significant.
During the 18-month study period, a total of 391 patients with migraine were interviewed regarding the semiology and the detailed clinical features of the attack. One hundred and sixty-six (42.5%) patients reported to be having neck pain during any time of the attack phase. Among them, 102 (61.5%) were female patients and the mean age of the study population was 35.8 ± 12.5 years (range, 14–65 years). Most patients were either housewives (35.3%) or students (26%), accounting for more than 61% of the study population. The median duration of headache at the time of presentation was 36 months. The maximum number of patients (45.3%) had 5–10 headache days per month during the initial 3 months of onset of headache; while during the time of the first hospital visit, the maximum number of patients (37.7%) had 20–25 headache days per month.
A total of 82.7% patients had migraine without aura. Among the patients presenting with migraine with aura, the most common form of the aura was visual aura (69.2%) followed by sensory aura (23%) and brainstem aura (7.6%). One patient had both visual aura followed by sensory aura. A total of 75.3% (n = 125) patients had episodic migraine at the time of onset, and in 48% (n = 60) of them, these episodes had gradually turned towards chronicity. Forty-one (24.6%) patients presented with features of chronic migraine at the time of onset; and, a total of 60.8% (n = 101) patients fulfilled the criteria of chronic migraine at the time of questionnaire filling in the hospital.
Among the 166 patients having neck pain during the attack phase, 53 patients (32%) had a history of neck pain, where it was thought to be a trigger for migraine. In the remaining 113 patients (68%), the neck pain was a part of migraine and absent in the non-headache phase. Thus, among the 166 patients having cervical migraine, only 53 (32%) had cervicogenic migraine.
Headache character and quality
In the study population, 54.2% patients had a typical unilateral side-shifting headache, 27.8% patients had side-locked headache (either restricted to right or left), and 18% patients had bilateral or holocranial type of headache [Figure 1]. The unilateral side-shifting character of migraine headache was more common in neck pain triggered migraineurs (69.8%) than in patients having neck pain as a part of migraine (47.7%), and the difference was statistically significant (P value = 0.025). In 75.2% of the cases, the headache was of throbbing or pulsating quality followed by bursting nature in 6.7% of the patients. Other patients rarely reported a band-like tightness around the neck sensation during their attacks. In 16.6% patients, headache was of a mixed quality [Figure 2]. However, the throbbing quality of pain was more common in patients having neck pain as a part of migraine (84%) than in those having neck pain triggered migraine (66%). This difference was statistically significant (P = 0.035). The pain intensity was mild in 6.5%, moderate in 51.2%, and severe in 42.3% patients. However, the intensity varied during different episodes of attacks in most patients.
Non-headache features of migraine
Among the non-headache features of migraine, photophobia was the most common symptom (94.6%) followed by nausea (91.5%) and phonophobia (88.5%). Other reported symptoms in the decreasing order of frequency were osmophobia, vertigo, vomiting, and cutaneous allodynia. Multiple non-headache symptoms (more than one ICHD-3-beta defined symptoms) were seen in 82.5% (n = 137) of the patients during their attacks. Thirty percent of the patients had four or more than four, nonheadache migrainous symptoms. This group included 5 patients who had all the above-mentioned seven symptoms during their attack. Seventy percent of the patients exhibited aggravation of pain during performance of physical activity. Cranial autonomic symptoms such as generalized or localized swelling in the scalp, temporal region or forehead, lacrimation, conjunctival injection, eyelid edema, nasal congestion, rhinorrhea, facial/forehead flushing, and facial/forehead sweating were seen in a significant number of our patients, i.e., 37.3% patients [Figure 3]. All patients had a headache duration lasting for more than 4 hours, and none of them had a short-lasting headache satisfying the criteria for diagnosing trigeminal autonomic cephalgia (TAC). There were no significant differences in the nonheadache symptoms, whether single or multiple (P = 0.587), and in the cranial autonomic symptoms (P = 0.596), between the neck pain triggered migraine patients and those having neck pain as a part of the attack. In the study population, 36 patients reported associated psychological or behavioral disturbances, which did not differ between the two groups of patients mentioned above (18.2% vs 24.7%; P = 0.08).
Cervical pain in different migraine attack phases
Among both the groups of patients, although 57 patients (34.3%) had neck pain that started in the prodromal phase, 44 patients continued to have neck pain during the migraine headache phase, and in the postdromal phase. One hundred and forty-eight patients (89.2%) experienced neck pain during the headache phase. Of these patients, 104 (62.6%) had their onset of headache just preceded by nuchalgia, or simultaneously accompanied by it. Five patients were reported to have the onset of neck pain during the postdromal phase, and a total of 46 patients (27.7%) had nuchalgia during this period. Sixteen (9.6%) patients gave a history of neck pain during all the phases of migraine attack [Figure 4]. There was no statistically significant difference between the two above-mentioned groups in relation to the occurrence of neck pain in different phases of migraine (P = 0.58).
Quality of neck pain
Whether neck pain triggered attack or the neck pain was as a part of migraine syptomatology, most patients (43.4%, n = 72) described stiffness or cramp-like neck pain during the migraine attack, followed by the throbbing type of pain in 26.5% (n = 44) and electric shock-like pain in 21.1% (n = 35) of the patients. Very few patients reported to be having a dull aching neck pain (7.2%, n = 12) or numbness in the upper neck region (1.8%, n = 3) during the attack phase [Figure 5]. Stiffness/cramp-like pain occurred more commonly in patients having neck pain that triggered their migraine attack, than with the other group. This difference was statistically significant (54.7% vs 35.3%; P < 0.001). A total of 95.2% patients described that their neck pain originated from the upper cervical region, and in 82.5% cases, the neck pain was unilateral. One hundred and twelve patients could detect the myofascial tenderness and painful neck movements during the migraine attack phase. Among them, 93 patients had myofascial tenderness and 65 patients had painful and restricted cervical movement at any time during the migraine phase. A total of 44 patients had cervical imaging (X-ray or magnetic resonance imaging) features of cervical spondylosis.
Age-wise distribution of patients with cervical migraine and its association with cervical spondylosis.
Forty-fourpatients (26.5% of total) had associated cervical spondylosis, that was evident both clinically and radiologically. Hence, to examine the association of cervical spondylosis in neck pain associated migraineurs, we divided the patients suffering from migraine into those with cervical migraine and cervical spondylosis. We also classified them according to their age groups (decade-wise). [Table 1] shows the age-wise distribution of patients with cervical migraine and their association with cervical spondylosis.
Medication overuse headache
Forty-three patients (26% of the total population) met the criteria for medication overuse headache. Most patients (n = 26, 60.5%) gave the history of a combination of analgesic overuse. Among the single analgesic overuse, the most common was non-steroidal anti-inflammatory drugs [NSAIDs] (the commonest being diclofenac, followed by naproxen or ibuprofen, followed by nimesulide or eterocoxib) in 9 patients, followed by paracetamol (n = 6), and aspirin (n = 2). In the combination group, tramadol and paracetamol combination drugs were more common (n = 15) than various NSAIDs and paracetamol (n = 11) combination drugs. No patients had a triptan overuse headache, although some patients had sometimes received triptans during the attack.
Our study, consisted of typical migraine patients (n = 391) of whom 42.5% (n = 166) were diagnosed to be having neck pain during the attack (cervical migraine). Among the patients with cervical migraine, approximately one-third (n = 53) of patients had a background history of neck pain, which was aggravated during the episode of migraine, where it was thought to be a trigger for the migraine attack. In the remaining 113 patients (68%), neck pain was a part of migraine as it subsided after the attack and was absent in the non-headache phase.
The non-headache migraine symptoms were seen in a significant proportion of cervical migraine patients with 80% of the patients having more than two symptoms, and 30% of the patients having four or more than four symptoms. Migraineurs with neck pain as a trigger, and patients having neck pain as a part of the attack had similar occurrence of non-headache migrainous symptoms (as the P value was not significant).
Most studies and the ICHD-3-beta state that migraine pain occurs commonly in the temporoparietal region, and less commonly in the occipital or orbitofacial location. However, the ICHD-3-beta did not mention neck pain as a site of migraine attack. Though neck pain is known to herald the onset of a migraine episode as a prodromal or premonitory symptom, its presence during the episode of migraine or as a trigger for the migraine attack is seldom described. In our study, 42.5% of migraine patients had neck pain anytime during the migraine attack. Among these patients, neck pain triggered the attack in one-third of the patients. This is in consonance with Wolff's comment that “even the neck” can be involved in migraine, and Selby's remark that “in severe attacks, pain may be felt in the neck.”, Moreover, neck pain occurred in our patients in different stages of migraine attack, which corroborates with other studies, suggesting that neck pain may be associated with any phase during the migraine attack.
There are limited studies that focus on the incidence, origin, or type of neck pain in migraine patients. In a study done in the early 1990s, 64% of the patients reported neck pain associated with the migraine attack, with 31% experiencing neck symptoms during the prodromal phase, 93% during the headache phase, and 31% during the recovery phase. In another study of 144 migraine patients, 75% reported neck pain associated with migraine attacks. Of these patients, 69% described their pain as “tightness”, 17% reported “stiffness”, and 5% reported “throbbing”. The neck pain occurred during the prodromal phase in 61%; during the headache phase in 92%, and during the recovery phase in 41% patients.
In a recent study, approximately 70% of migraineurs reported neck pain at any time during the migraine phase. Among them, 54.4% noticed neck pain with the start of the headache phase; 24.2% reported neck pain within 2 hours before the headache phase and 7.4% experienced neck pain 2–48 hours before the headache phase. Thus, most studies mentioned that 60–70% migraine patients may have neck pain during the attack. However, in this study, we noticed a slightly lower percentage (42.5%) of migraineurs having neck pain at any time during migraine attack. This was probably because we considered patients as having neck pain only when it was consistently felt during at least five migraine attacks.
Our study is in accordance with the above-mentioned studies concerning the association of neck pain in different phases of migraine. Most authors have described that in more than 90% migraineurs, neck pain was felt at the onset of the headache phase. We also found that 89.2% patients had neck pain at the onset of migraine headache. This view suggests that neck pain may not only be a premonitory symptom. Rather, it is a part of the migraine symptom complex (occurring in the headache phase), and in some patients (32% of total) it may also trigger the attack, thus creating a new concept of cervical migraine. Most of our patients were housewives or students. The explanation for their cervical origin of migraine may be that the neck was held in one position for a long period, while they were cooking or studying, as is a common practice in the Indian population [Table 2].
There are certain other specific features that we observed. These are not mentioned in other studies. An unusually high number (60.8%) of our cervical migraine patients developed chronicity (more than 15 headache days per month) within 3 years. Unilateral side shifting headache, which is commonly seen in a significant population of migraine patients, was observed only in approximately half of our cervical migraine patients. Throbbing or pulsating quality of headache, which was invariably seen in all migraineurs and is one of the essential components of the diagnostic criteria in migraine patients, was seen in only three-fourth of our cervical migraine patients with another one-sixth of them having a mixed quality of headache. Hence, the throbbing character and side-shifting nature were less frequent among the cervical migraineurs in our study.
Although a majority of our patients, 43.4% (n = 72) described stiffness or cramp-like neck pain during the migraine attack, many patients also experienced neck myofascial tenderness (n = 93) and painful cervical movement (n = 65) during the attack. Some patients described the neck movements as an aggravating factor that also contributed to the severity of headache, particularly among the cohort of housewives and students.
Cranial autonomic dysfunction, which was thought to be a cardinal feature of trigeminal autonomic cephalgia, may also be observed in migraine patients with varying frequency (possibly in up to 50% of the patients)., In our study, 37.3% patents had one or more than one cranial autonomic symptoms and all had headache for a duration of more than 4 hours. None of our patients, therefore, satisfied the criteria for autonomic cephalgia. Many patients described a localized swelling or bogginess in the temporal region. Some had forehead or scalp swelling, which may be attributed to the autonomic symptoms. Most symptoms were bilateral, and ocular symptoms (lacrimation, conjunctival injection, or eyelid edema) were the most common ones present, followed by forehead/facial (flushing or sweating) and nasal symptoms (congestion or rhinorrhea). A single symptom was more common than a combination of symptoms, and none of our patients were reported to be having aural symptoms.
Forty-four patients (26.5%) had clinical features and radiological findings suggestive of cervical spondylosis. Whether this presence contributed to the neck pain in the migraine attack or was just an association cannot be unequivocally established. We, however, studied the incidence of neck pain in migraine patients stratified according to different age groups. As shown in [Table 2], neck pain among the migraineurs did not vary significantly in the different age groups. Neck pain occurred either as a trigger or as a part of migraine headache in 39–46% of patients in different groups. Another important observation was that the prevalence of cervical spondylosis increased with age (2.7% among the 14–25-year age group, to 60% among the 56–66-year age group patients), which did not significantly affect the incidence of neck pain among the cervical migraine patients. Specifically, the presence of cervical spondylosis may just be an association, and in all likelihood, did not contribute to the occurrence of neck pain in migraine patients [Figure 6].
Several patients (22.3%) fulfilled the criteria of medication overuse headache, among the patients presenting with chronic headache, i.e., an average of one-third chronic migraine patients in this study population had medication overuse headache and most of them received a combination of analgesic medications prescribed over the counter, with tramadol and paracetamol combination being the commonest. The prevalence of medication overuse headache in chronic headache patients may range between 30% and 50% according to the literature.
Fifty-seven patients (34.3%) suffering from cervical migraine had an associated tension-type headache. The figure suggested that neck pain may be associated with both migraine and tension-type headache as the major extracranial symptom. This was despite the fact that migraine (trigeminovascular mechanism) and tension-type headache (muscular-tension mechanism) had differing mechanisms of origin. The exact association of cervical pain in both these types of primary headache needs further investigation.
In our study, neck pain-triggered migrainous headache was evidenced by the following: (1) the maximum number of patients had neck pain just before the onset of headache phase and most of them had disappearance of neck pain after the headache phase, (2) the neck pain radiated to the frontotemporal or the orbitofacial regions in many patients, with a significant number of patients also having non-headache migrainous symptoms, (3) many patients had ipsilateral neck pain in consonance with the laterality of headache, (4) the attacks were often precipitated either by neck movements or by pressure on the myofascial tender spots in the neck, and (5) in some patients, there was background neck pain, which got aggravated during the migraine attack, suggesting the fact that neck pain at times triggers the headache. We also ruled out significant secondary pathologies of the neck or cervical spine, which may cause severe secondary headache that may serve to obscure the quality of the typical migrainous headache and neck pain.
The association of neck pain and migraine may result from various pathophysiological mechanisms. No studies have been carried out to show the possible causal association of neck pain and migraine The neck structures innervated by the first three cervical nerves can be associated with migraine through the convergence of nociceptive afferents at the level of the caudal part of the trigeminal nucleus in the brainstem and the sensitization of the trigeminocervical complex., Very often, authors have thought that the neck pain is a prodromal symptom or is regarded as an extracranial manifestation of migraine if it is associated with headache. However, patients generally report that their migraine “starts” in the neck often stating that their neck is the “cause” of their migraine. Hence, many researchers are studying a possible cervical origin of migraine. Studies have mentioned that neck pain more commonly starts with migrainous headache rather than being a prodromal manifestation, and a significant number of patients had this symptomatology without any cervical abnormality., These studies have also proposed certain pathophysiological mechanisms, which have not been clearly defined. Prolonged nociceptive stimuli from the neck structures could also be an important trigger for producing continuous afferent bombardment of the trigeminal nerve nucleus caudalis, and hence, activation of the trigeminovascular system., As the pathogenesis of migraine is linked to the trigeminal innervations of the cranial blood vessels, noxious stimuli from the cervical structures may also play a role in this pathogenesis by facilitating central sensitization.
The major limitation of this study was that it was partly conducted using memory-based interviews, possibly resulting in a patient selection bias. In many migraineurs, the subtypes of migraine and the associated symptoms may vary from attack to attack. Therefore, sometimes the patients' version may have required the examiners' interpretation for documentation, leading to an inherent bias. In certain patients, psychological factors such as anxiety and depression, which may have a modulating effect on pain perception could not be eliminated from our study.,, While anxiety and depression are associated with migraine, these may also cause neck pain and headache. However, in our study the role of anxiety and depression as a contributory factor for neck pain in our patients suffering from migraine, has not been evaluated. Again, it was unclear from the study whether only functional cervical pain was associated with migraine, or a structural cervical pain (when associated with a neck pathology such as cervical spondylosis) might have also triggered the migraine attack.
This study highlights that a significant proportion of patients had neck pain during the migraine attack and most of them had the onset of neck pain during the headache phase. Although not conclusive, the above study reveals that a majority of patients may have cervical pain as a part of migraine headache, and possibly, the neck pain acts as a trigger for the attack. Precipitation of migraine in some patients after keeping the neck in one posture for a long time may suggest cervical mechanisms as the triggers for the central mechanisms of migraine. On the other hand, the presence of neck pain during the headache phase would suggest that it could also be a peripheral manifestation of migraine. This theory is supported by studies showing that nociceptive afferents from the meningovascular structures and the upper three cervical nerves converge to the same second order neurons in the trigeminocervical complex. This clinical observation, however, requires experimental confirmation. Neck pain may result in increased disability in headache sufferers that may have a considerable social impact. In conclusion, neck pain triggered headache is not always cervicogenic headache or secondary headache (headache due to cervical pathology), but in a majority of patients, it may be a manifestation of cervical migraine (migraine with cervical pain as a trigger or a part of the attack). Both these two entities have similar presentations and symptomatology and require careful clinical interpretation. Hence, the physician and neurologist should be aware of this fact and take a thorough history in the case of neck pain triggered and/or associated headache to rule out the presence of cervical migraine. This will be helpful in avoiding unnecessary imaging or investigation.
We thank Mr. Shakti Kumar for his secretarial help.
Financial support and sponsorship
This study was approved by Institutional Ethics Committee, SGPGIMS, Lucknow, India.
Conflicts of interest
The authors declare no conflict of interest issues.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]
[Table 1], [Table 2]